Imaging display method, apparatus, and storage medium based on security inspection machine

Abstract

The application discloses a kind of based on the imaging display method, equipment and storage medium of security inspection machine, the imaging display method based on the security inspection machine includes: the complete package recognition processing is carried out to current buffer image obtained, and complete package recognition result is obtained;In response to complete package recognition result Characterization exists in current buffer image complete inspection package, the target detection processing is carried out to the package image area where complete inspection package is located, and target detection result is obtained;In response to target detection result Characterization exists in complete inspection package target object, then target object is displayed processing.The above scheme can reduce the security inspection pressure of security guard.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of security inspection machines, in particular to an imaging display method and device based on a security inspection machine and a storage medium. BACKGROUND

[0002] A security inspection machine, also known as a security inspection instrument, commonly includes a security inspection X-ray machine, a baggage security inspection machine, a channel type X-ray machine, and a material inspection X-ray machine. It is widely used in the logistics industry, public transport hubs, and various important places.

[0003] For example, in the logistics industry, as the logistics industry develops and progresses, the volume of goods is increasing, and logistics security inspection is becoming increasingly important. In order to improve the speed of security inspection, the running speed of the conveyor belt of a high-speed security inspection machine is very fast, resulting in an increasingly fast imaging speed.

[0004] However, in the existing imaging method of a high-speed security inspection machine, due to the increased speed of the conveyor belt and the imaging speed, the display time of the package after imaging is very short, that is, the package will be quickly brushed on the display interface, thereby increasing the security inspection pressure of the security inspector. SUMMARY

[0005] The present application provides at least an imaging display method, device, equipment, and computer readable storage medium based on a security inspection machine.

[0006] The first aspect of the present application provides an imaging display method based on a security inspection machine, comprising: performing complete package identification processing on a current buffer image obtained, to obtain a complete package identification result; in response to the complete package identification result representing that there is a complete package to be inspected in the current buffer image, performing target detection processing on a package image region where the complete package to be inspected is located, to obtain a target detection result; and in response to the target detection result representing that there is a target object in the complete package to be inspected, performing display processing on the target object.

[0007] In an embodiment, after the complete package identification processing on the current buffer image obtained is performed to obtain the complete package identification result, the method further comprises: in response to the complete package identification result representing that there is an incomplete package to be inspected in the current buffer image, performing rendering imaging processing on current radiation data obtained continuously, to obtain a current rendering image; performing buffer update processing on the current buffer image according to the current rendering image, until the complete package identification result of the updated current buffer image represents that there is the complete package to be inspected in the updated current buffer image; and performing target detection processing on the package image region where the complete package to be inspected is located, to obtain the target detection result.

[0008] In an embodiment, the current buffer image is cached in a buffer area, and the buffer updating processing of the current buffer image according to the current rendering image comprises: caching the current rendering image in the buffer area; and updating the current buffer image according to the current rendering image in the buffer area to obtain the updated current buffer image.

[0009] In an embodiment, the target detection processing of the package image area where the complete to-be-inspected package is located to obtain a target detection result comprises: obtaining preset image features of each target object; performing image feature extraction processing on the complete to-be-inspected package to obtain to-be-inspected features; determining target features with feature similarity greater than a similarity threshold from each preset feature according to the feature similarity between the to-be-inspected features and each preset feature; in response to the number of target features being non-zero, obtaining a target detection result representing that the target object exists in the complete to-be-inspected package; and in response to the number of target features being zero, obtaining a target detection result representing that the target object does not exist in the complete to-be-inspected package.

[0010] In an embodiment, the display processing of the target object comprises: performing cutout processing on the current buffer image according to the position information of the target object in the current buffer image to obtain a target object image; performing image optimization processing on the target object image to obtain an optimized image; and performing display processing on the optimized image.

[0011] In an embodiment, the image optimization processing of the target object image to obtain an optimized image comprises: obtaining a material type of the target object; in response to the material type of the target object being inorganic matter, determining organic matter in the target object image according to corresponding ray data of the target object image; and performing elimination processing on the organic matter in the target object image to obtain the optimized image.

[0012] In an embodiment, after the complete package identification processing of the obtained current buffer image to obtain a complete package identification result, the method further comprises: in response to the complete package identification result representing that there is an incomplete to-be-inspected package in the current buffer image, performing image feature extraction processing on the incomplete to-be-inspected package to obtain local features; performing target prediction processing according to the local features to obtain a target prediction result; and performing display processing on the target prediction result.

[0013] In an embodiment, the method further comprises: determining an occurrence frequency of the target object in a preset period according to target detection results in the preset period; and adjusting a track speed of the security inspection machine according to the occurrence frequency, the occurrence frequency being negatively correlated with the track speed.

[0014] The second aspect of the present application provides an imaging display device based on a security inspection machine, comprising: a package recognition module, configured to perform complete package recognition processing on an obtained current buffer image to obtain a complete package recognition result; a target detection module, configured to, in response to the complete package recognition result representing that there is a complete to-be-inspected package in the current buffer image, perform target detection processing on a package image region where the complete to-be-inspected package is located to obtain a target detection result; and a target display module, configured to, in response to the target detection result representing that there is a target object in the complete to-be-inspected package, perform display processing on the target object.

[0015] The third aspect of the present application provides an electronic device, comprising a memory and a processor, the processor being configured to execute program instructions stored in the memory to implement the above-mentioned imaging display method based on a security inspection machine.

[0016] The fourth aspect of the present application provides a computer-readable storage medium, having program instructions stored thereon, the program instructions being executed by a processor to implement the above-mentioned imaging display method based on a security inspection machine.

[0017] In the prior art, when a security inspection machine scans a package, the current scanned ray data is rendered and displayed in real time. Real-time image pushing and fast moving display can cause eye fatigue. The method of the present application buffers the image rendered by the scanned ray data when the security inspection machine scans the package, performs complete package recognition processing on the obtained current buffer image to obtain a complete package recognition result, and thereby determines whether complete package information exists in the current buffer image. If the complete package recognition result represents that there is a complete to-be-inspected package in the current buffer image, target detection processing is performed on the package image region where the complete to-be-inspected package is located in the current buffer image to obtain a target detection result. Thereby, it is determined whether the complete to-be-inspected package has a target object that needs to be focused on. In response to the target detection result representing that there is a target object in the complete to-be-inspected package, display processing is performed on the target object. Thereby, the image pushing mode can be adjusted from one or more groups of ray data to complete package data pushing, the real-time pushing process is omitted, and eye fatigue is reduced. Moreover, the package with the target object is displayed again, the number of images that need to be judged by the security inspector is reduced, and the security inspection pressure is reduced.

[0018] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, but not limiting the present application. BRIEF DESCRIPTION OF DRAWINGS

[0019] The drawings incorporated into the specification and forming part of the specification, show embodiments consistent with the present application, and together with the specification, serve to explain the technical solutions of the present application.

[0020] Figure 1 is a flowchart of an exemplary embodiment of the imaging display method based on a security inspection machine of the present application;

[0021] Figure 2 is a schematic diagram of an exemplary image optimization process in the imaging display method based on a security inspection machine of the present application;

[0022] Figure 3 is an exemplary application flowchart in the imaging display method based on a security inspection machine of the present application;

[0023] Figure 4 is a block diagram of the imaging display device based on a security inspection machine shown in an exemplary embodiment of the present application;

[0024] Figure 5 is a structural schematic diagram of an embodiment of the electronic device of the present application;

[0025] Figure 6 is a structural schematic diagram of an embodiment of the computer readable storage medium of the present application. DETAILED DESCRIPTION

[0026] The scheme of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0027] In the following description, specific details are set forth in order to provide a thorough understanding of the present application. The present application can be practiced without resorting to the details specifically set forth.

[0028] The term "and / or" herein merely describes an associated relationship between associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects. In addition, "multiple" herein means two or more. In addition, the term "at least one" herein means any one of multiple or any combination of at least two of multiple, for example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0029] For ease of understanding, one of the applicable scenarios of the present application will now be described by way of example. The imaging principle of the security inspection machine is mainly based on the penetration characteristics of X-rays and the difference in the degree of X-ray absorption by matter. The X-rays are converted into electrical signals by a detector and processed into images, and the density and material of the object are displayed in different colors or gray scales.

[0030] The imaging process of existing security screening machines is typically real-time scanning. For example, when a package enters the security screening channel, it triggers relevant sensors, and an X-ray source emits an X-ray beam that penetrates the moving item on the machine's conveyor belt, thus achieving line-by-line scanning. The penetrating X-rays are captured by detectors and converted into electrical signals. These electrical signals are then amplified, filtered, and digitized to generate image signals, which are finally displayed line by line as an image of the package.

[0031] With the widespread use of high-speed security inspection machines, the conveyor belts of these machines are moving at increasingly faster speeds, resulting in faster image display speeds. This negatively impacts the ability to clearly see and confirm the presence of target objects within packages. While current image display methods have shifted from displaying a few lines of ray data per frame to combining several lines of ray data per frame, reducing image flicker during security checks, the increased imaging speed in high-speed machines means that the display time for the last package on the current screen is very short, hindering manual image interpretation by security personnel. Furthermore, the high-speed movement of the machine's conveyor belt rapidly refreshes the screen with packages, resulting in very short package imaging time and placing significant pressure on security personnel for image interpretation.

[0032] Please see Figure 1 , Figure 1 This is a schematic flowchart of an exemplary embodiment of the imaging display method based on a security inspection machine according to this application. Specifically, it may include the following steps:

[0033] Step S110: Perform complete package recognition processing on the acquired current buffer image to obtain the complete package recognition result.

[0034] The current buffer image refers to the image temporarily stored in the preset buffer. The X-ray source of the security scanner emits radiation energy, which, after passing through the security checkpoint, hits the crystal on the detector plate to obtain radiation data. The radiation data is then rendered into an image according to a preset rendering algorithm, and the rendered image is placed into a loopable buffer.

[0035] The complete package recognition process is mainly used to determine whether the image content of the current buffered image in the buffer constitutes a complete package. If the current buffered image does not represent an image containing a complete package, the rendered image needs to be stored in the buffer until a complete package is recognized in the current buffered image. This application does not limit the shape, size, etc. of the package.

[0036] It should be noted that the process of obtaining the current buffered image for complete parcel identification processing in the present application can be periodic, or real-time, or in response to a predetermined condition (for example, every time one or more new images are stored in the buffer, the current buffered image in the buffer is processed for complete parcel identification), which is not limited here.

[0037] Exemplarily, the method of complete parcel identification processing can include but is not limited to obtaining the image features (such as histogram features, geometric features, etc.) of the current buffered image, inputting the image features into the pre-trained complete parcel identification model (the complete parcel identification model can be trained based on a neural network classification model), and thereby outputting the parcel type (such as complete parcel, incomplete parcel, etc.) corresponding to the image features through the complete parcel identification model. The specific implementation method can refer to the related technical principles, which will not be repeated here.

[0038] Step S120, in response to the complete parcel identification result indicating that there is a complete parcel to be inspected in the current buffered image, performing target detection processing on the parcel image region where the complete parcel to be inspected is located, and obtaining a target detection result.

[0039] In combination with the foregoing steps, if the complete parcel identification result indicates that there is a complete parcel to be inspected in the current buffered image, target detection processing can be performed on the parcel image region where the complete parcel to be inspected is located in the current buffered image to determine whether there is a target object in the complete parcel to be inspected.

[0040] Exemplarily, the method of target detection processing can include but is not limited to obtaining the image features (inspection features) of the complete parcel to be inspected. By comparing the inspection features with the reference features in the image feature library to obtain the feature similarity, the reference features with a feature similarity greater than a similarity threshold can be determined as target features, and the target object corresponding to the target features can be determined as the target object corresponding to the inspection features. The specific implementation method can refer to the related technical principles, which will not be repeated here.

[0041] Another exemplary, based on the judgment according to the feature similarity between the inspection features and the reference features, the judgment can also be combined with the features of the parcel owner. For example, the parcel of a construction worker may contain construction tools, so if the target object to be detected is a construction tool, the similarity threshold can be reduced for the parcel of the construction worker to improve the detection sensitivity of the detection module to the construction tool. Wherein, the determination of the ownership relationship between the current analyzed complete parcel to be inspected and the parcel owner can refer to the existing related technology, which will not be repeated here. For example, when the parcel owner puts the parcel into the security inspection machine, the parcel owner id and the parcel id can be generated, and the ids can be bound to each other to determine the parcel owner corresponding to the parcel during security inspection.

[0042] Further, the characteristics of the parcel owner can include, but are not limited to, appearance characteristics, behavior characteristics, gender characteristics, and / or occupation characteristics, etc. For the characteristic detection of the parcel owner, a corresponding personnel characteristic library can also be constructed in advance. If the characteristics of the parcel owner match the reference characteristics in the personnel characteristic library (the similarity is greater than the personnel similarity threshold), the similarity threshold that needs to be referred to in the target detection process can be reduced, or the similarity threshold corresponding to the matched reference characteristics can be reduced (different types of parcel owners need to focus on different types of target objects, so the similarity threshold when judging the corresponding target object can be reduced for the type of the current parcel owner). The specific implementation method can also be adjusted and changed according to the related technical principles, which will not be repeated here.

[0043] Step S130, in response to the target detection result representing that the target object exists in the complete parcel to be inspected, the target object is displayed.

[0044] The target object refers to the object that needs to be detected and focused on. The target object can be one or more, which can be living objects and / or inanimate objects, etc., which are not limited here. One complete parcel to be inspected can have one or more target objects, or no target object.

[0045] In combination with the foregoing steps, if the target detection result represents that the target object exists in the complete parcel to be inspected, the target object needs to be displayed. The method of display processing can include, but is not limited to, displaying the current buffer image where the target object exists, displaying the parcel image of the complete parcel to be inspected where the target object exists, displaying the target object, or displaying the target library image of the same type as the target object, etc., which are not limited here. If the target detection result represents that the target object does not exist in the complete parcel to be inspected, the image of the complete parcel to be inspected can be selected not to be displayed.

[0046] It can be seen that, by buffering the image rendered by the scanned ray data when the security inspection machine scans the package, the complete package recognition result is obtained by performing complete package recognition processing on the obtained current buffer image, thereby judging whether complete package information exists in the current buffer image. If the complete package recognition result represents that complete inspection package exists in the current buffer image, target detection processing is performed on the package image region where the complete inspection package is located in the current buffer image to obtain a target detection result. Thereby, it is judged whether the complete inspection package has a target object that needs to be focused on. In response to the target detection result representing that the target object exists in the complete inspection package, display processing is performed on the target object. Compared with the prior art, when the security inspection machine scans the package, the currently scanned ray data is rendered and displayed in real time, and the real-time image pushing and fast moving display will cause eye fatigue. The method of the present application can adjust the data pushing mode from one or more sets of ray data pushing to complete package data pushing, omit the real-time pushing process, and reduce eye fatigue. Moreover, the method of the present application displays the package with the target object, reduces the number of images that need to be judged by the security inspector, and reduces the security inspection pressure.

[0047] On the basis of the above-mentioned embodiments, the steps after the complete package recognition processing on the obtained current buffer image to obtain the complete package recognition result are described in the embodiments of the present application. Specifically, the method of the present embodiment comprises the following steps:

[0048] In response to the complete package recognition result representing that the incomplete inspection package exists in the current buffer image, the current ray data is continuously obtained, and rendering imaging processing is performed on the current ray data to obtain a current rendering image; the current buffer image is updated according to the current rendering image until the complete package recognition result of the updated current buffer image represents that the complete inspection package exists in the updated current buffer image; and target detection processing is performed on the package image region where the complete inspection package is located to obtain a target detection result.

[0049] In combination with the foregoing embodiments, if the complete package recognition result represents that the incomplete inspection package exists in the current buffer image, that is, the security inspection machine has not completed scanning the currently detected package. Therefore, the current ray data is continuously obtained, and the current ray data is rendered and imaged according to a preset rendering algorithm to obtain a current rendering image. Then, the current rendering image is stored in the buffer, and the previously stored image (current buffer image) in the buffer is updated. Thereby, the cycle is repeated until the complete package recognition result of the updated current buffer image represents that the complete inspection package exists in the updated current buffer image, that is, the security inspection machine has completed scanning the currently detected package. Then, target detection processing is performed on the package image region where the complete inspection package is located to obtain a target detection result.

[0050] On the basis of the above-mentioned embodiments, the embodiments of the present application illustrate the step of performing buffer updating processing on the current buffer image according to the current rendering image. The current buffer image is buffered in the buffer area. Specifically, the method of the embodiments comprises the following steps:

[0051] buffering the current rendering image to the buffer area; performing updating processing on the current buffer image according to the current rendering image in the buffer area to obtain the updated current buffer image.

[0052] In combination with the foregoing embodiments, the current buffer image is buffered in the buffer area buffer. In the process of performing buffer updating on the current buffer image according to the image rendered by the newly obtained ray data (current rendering image), the current rendering image is also buffered in the same buffer area, so that the current rendering image and the current buffer image form a new image in the buffer area, that is, the updated current buffer image is obtained.

[0053] It should be noted that the complete wrapping recognition result obtained by performing complete wrapping recognition processing on the current buffer image after a certain update does not necessarily represent that it contains complete wrapping. If there is no complete wrapping, the ray data is continuously obtained for rendering and imaging and stored in the buffer area by analogy with the foregoing examples. If there is complete wrapping, the current buffer image in the buffer area can be read for target detection processing, and the buffered image temporarily stored in the buffer area is emptied, and then the image rendered by the ray data for the new package can be obtained for temporary storage.

[0054] On the basis of the above-mentioned embodiments, the embodiments of the present application exemplarily illustrate the step of performing target detection processing on the package image area where the complete package to be detected is located to obtain the target detection result. Specifically, the method of the embodiments comprises the following steps:

[0055] obtaining preset image features of each target object; performing image feature extraction processing on the complete package to be detected to obtain a detection feature; determining a target feature with a feature similarity greater than a similarity threshold from each preset feature according to the feature similarity between the detection feature and each preset feature; in response to the number of target features being non-zero, obtaining a target detection result representing that there is a target object in the complete package to be detected; and in response to the number of target features being zero, obtaining a target detection result representing that there is no target object in the complete package to be detected.

[0056] In combination with the foregoing embodiments, in the target detection on the complete parcel to be inspected, preset image features of the target object to be detected can be acquired first. The preset image features can be pre-stored in an image feature library and can be used as reference features in the target detection. Thus, the image features of the complete parcel to be inspected can be extracted to obtain the inspection features. Based on the feature similarity between the inspection features and each preset image feature, the target features similar to the inspection features (i.e., the preset image features with the feature similarity greater than a similarity threshold) are determined, and the target object represented by the target features is determined as the target object corresponding to the inspection features.

[0057] If the number of target features is not zero, the target detection result indicating that the target object exists in the complete parcel to be inspected is obtained. If the number of target features is zero, the target detection result indicating that the target object does not exist in the complete parcel to be inspected is obtained, that is, the image of the parcel can not be displayed.

[0058] Optionally, the reference image of the target object to be detected can also be pre-stored in the image library, and the preset image features of the same target object correspond to the reference image, or the preset image features of the same type of target object correspond to the reference image. Thus, after the target features similar to the inspection features are determined, the corresponding target reference image can be searched in the reference images according to the target features, and the target reference image is displayed. The image format of the reference image is not limited in the present application. For example, the reference image can be an RGB format image, and the target reference image in the RGB format is displayed, which is beneficial to the security officer to quickly distinguish the material type of the target object represented by the image, and compared with the image directly displayed after the ray data is rendered in the traditional security inspection machine imaging display method, the material type of the target object can be more clearly represented.

[0059] On the basis of the foregoing embodiments, the present embodiment exemplarily illustrates the steps of displaying the target object. Specifically, the method of the present embodiment comprises the following steps:

[0060] According to the position information of the target object in the current buffer image, the current buffer image is subjected to a cutout processing to obtain a target object image, the target object image is subjected to an image optimization processing to obtain an optimized image, and the optimized image is subjected to a display processing.

[0061] With the foregoing embodiments, in order to highlight the target object in the display interface, the current buffer image can be subjected to a cutout processing according to the position information (such as pixel coordinates, etc.) of the target object in the current buffer image, to obtain a target object image. Other image regions outside the target object image can be set to a specified color (such as white, etc.). Then the target object image can be subjected to an image optimization processing to obtain an optimized image. The optimized image is then displayed. The image optimization processing can include, but is not limited to, noise reduction, sharpening, contrast adjustment, color adjustment, etc.

[0062] Based on the foregoing embodiments, the embodiments of the present application exemplarily illustrate the step of subjecting the target object image to an image optimization processing to obtain an optimized image. Specifically, the method of the present embodiment includes the following steps:

[0063] The material type of the target object is obtained; in response to the material type of the target object being inorganic matter, the organic matter in the target object image is determined according to the ray data corresponding to the target object image; the organic matter in the target object image is subjected to a rejection processing to obtain an optimized image.

[0064] With the foregoing embodiments, in the image optimization processing of the present application, a corresponding image optimization method can be selected according to the material type of the target object.

[0065] It can be understood that, since different materials (such as metal, liquid, organic matter, etc.) have different absorption degrees of X-rays, the material type of the target object is determined according to the ray data of the security inspection instrument. In specific application scenarios, the target object to be detected can be of various material types, which are not limited herein.

[0066] Exemplarily, if the material type of the target object is a first material, the other materials in the target object image except the first material can be determined according to the ray data corresponding to the target object image; the image regions corresponding to the other materials can be subjected to a rejection processing or a transparent processing (setting the transparency of this part of the image regions).

[0067] For example, as shown in FIG. 1, Figure 2 Figure 2 ​is an exemplary image optimization process schematic diagram in the imaging display method based on the security inspection machine of the present application. If the material type of the target object is inorganic matter (for example, metal appliances), the organic matter (for example, plastic shell or leather shell) in the target object image is determined according to the ray data corresponding to the target object image; the organic matter in the target object image is removed to obtain an optimized image. Optionally, the image of the inorganic matter can also be enhanced and displayed during the image optimization process. Thus, the identification interference of other material objects in the target object image can be reduced, the appearance of the target object can be more intuitively presented, and the type and corresponding probability of the target object can also be displayed. In the subsequent display process, the target object can be displayed alone or together with the package in which the target object is located, which is not limited here.

[0068] On the basis of the above-mentioned embodiments, the present embodiment describes the steps after the complete package recognition result is obtained by performing complete package recognition processing on the acquired current buffer image. Specifically, the method of the present embodiment includes the following steps:

[0069] In response to the complete package recognition result representing that there is an incomplete target package in the current buffer image, image feature extraction processing is performed on the incomplete target package to obtain local features; target prediction processing is performed according to the local features to obtain a target prediction result; and the target prediction result is displayed.

[0070] In combination with the foregoing embodiments, the foregoing embodiments mainly describe the method of target detection when it is recognized that there is a complete target package in the current buffer image. Then, the detected target object is displayed, so that the security inspector does not need to continuously observe the image displayed on the screen as in the conventional method.

[0071] The local features refer to the features extracted from the incomplete target package (i.e., the local features in the complete target package).

[0072] The present embodiment provides a method of predicting whether there is a target object in the package based on the features (local features) of the package already displayed in the current buffer image by target prediction when it is recognized that there is an incomplete target package in the current buffer image.

[0073] Exemplarily, if it is determined that there is an incomplete target parcel in the current buffered image stored in the buffer, image features (local features) of the incomplete target parcel can be extracted. Then, target prediction processing is performed based on the extracted local features to obtain a target prediction result, and the target prediction result is displayed. The target prediction processing can be performed by using a traditional machine learning model or a deep learning model, and details are not described herein. The target prediction processing can be performed periodically, in real time, or in response to a preset trigger condition (for example, the local features of the current buffered image obtained in an initial period of time cannot accurately predict the corresponding target object. Therefore, the target prediction processing can be performed after a rendering image corresponding to N ray data is obtained, or after M frames of rendering images are obtained. N needs to be greater than a preset threshold of the number of rays, and M needs to be greater than a preset threshold of the number of frames. In this way, the computing resources in the initial period of time can be saved, and the waste of computing resources in the period with low prediction probability can be avoided.

[0074] It should be noted that as the number of current buffered images in the buffer increases, and there is still no complete target parcel in the current buffered images, the feature information contained in the local features extracted from the current buffered images will be more sufficient. Therefore, the target prediction result can change with the number of current buffered images, and details are not described herein.

[0075] Further, for the process of displaying the target prediction result, the target prediction result can include zero, one or more target objects (for the sake of distinction, referred to as predicted objects) corresponding to the local features, and a prediction probability corresponding to each predicted object. If there is one predicted object, the related information of the predicted object (such as the name of the predicted object, the corresponding base image, and / or the corresponding prediction probability) can be displayed. If there are multiple predicted objects, the predicted objects can be sorted according to the prediction probability, and the related information of the predicted object with the highest prediction probability or the predicted objects with higher prediction probability (for example, the top A predicted objects with higher probability after sorting) can be displayed. Or, without sorting, the related information of the predicted objects with a prediction probability greater than a prediction probability threshold can be displayed. Alternatively, the predicted objects with a prediction probability greater than a prediction probability threshold are first selected; if the number of selected predicted objects is one, the selected predicted objects are displayed; if the number of selected predicted objects is multiple, the selected predicted objects are sorted and displayed. The specific sorting algorithm is not limited herein, and can be set according to the actual application scenario. In this way, a certain prediction time can be provided to the security personnel before the complete target parcel is displayed, so that the security personnel can more quickly react and determine the target object in the complete target parcel when observing the complete target parcel.

[0076] In addition, if the local feature corresponds to zero target object or the prediction probability between the local feature and each target object is less than the prediction probability threshold in the target prediction processing process, the target prediction result can be selected not to be displayed.

[0077] On the basis of the above-mentioned embodiments, the application can also dynamically adjust the speed of the conveyor belt of the security inspection machine according to the target detection result. Specifically, the method of the embodiment comprises the following steps:

[0078] According to the target detection result in the preset period, the frequency of occurrence of the target object in the preset period is determined; and the speed of the conveyor belt of the security inspection machine is adjusted according to the frequency of occurrence, and the frequency of occurrence is negatively correlated with the speed of the conveyor belt.

[0079] In combination with the foregoing embodiments, the frequency of occurrence of the target object in the preset period can be counted according to the target detection result in the preset period. If the frequency of occurrence of the target object is higher, the speed of the conveyor belt of the security inspection machine needs to be reduced to provide sufficient security inspection time for the security inspector. If the frequency of occurrence of the target object is lower, the speed of the conveyor belt of the security inspection machine can be increased to improve the security inspection efficiency. The specific adjustment strategy is not limited here, and can be set as needed according to the actual application scene.

[0080] On the basis of the above-mentioned embodiments, the embodiment illustrates an example application scene of the imaging display method based on the security inspection machine of the application. As shown in FIG. 8, Figure 3 Figure 3 is an example application flowchart of the imaging display method based on the security inspection machine of the application. After the security inspection process starts, the ray energy data on the detection plate crystal after the X-ray source of the security inspection machine passes through the channel is acquired. The acquired ray energy data is transmitted to the rendering algorithm for rendering imaging. The image of the rendering imaging is put into the loop buffer. The content of the image in the buffer is determined whether it is a complete package by the complete package recognition algorithm. If it is a complete package, target detection is performed on the package to determine whether there is a target object in the package. At the same time, the previously stored images in the buffer can be emptied, and the next package image can be cached. If it is not a complete package, the acquired ray data is continuously rendered and imaged and stored in the buffer. If there is a target object in the complete package, the package can be displayed through the display interface, or the target object can be extracted from the package and displayed through the display interface. If there is no target object in the complete package, no display is performed. Subsequently, whether there is a target object in the package or not, the image of the package can be stored in the hard disk for persistence, so as to facilitate image query and other application processes.

[0081] ​It should be further explained that the execution subject of the imaging display method based on the security inspection machine can be an imaging display device based on the security inspection machine, for example, the imaging display method based on the security inspection machine can be executed by a terminal device or a server or other processing device, wherein the terminal device can be a user equipment (UE), a computer, a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementation manners, the imaging display method based on the security inspection machine can be realized by a processor calling computer readable instructions stored in a memory.

[0082] Figure 4 is a block diagram of an imaging display device based on a security inspection machine shown in an exemplary embodiment of the present application. As shown in the figure, the exemplary imaging display device based on the security inspection machine 400 includes a package identification module 410, a target detection module 420 and a target display module 430. Specifically: Figure 4

[0083] The package identification module 410 is configured to perform complete package identification processing on the acquired current buffer image to obtain a complete package identification result.

[0084] The target detection module 420 is configured to, in response to the complete package identification result representing that there is a complete target package in the current buffer image, perform target detection processing on a package image region where the complete target package is located to obtain a target detection result.

[0085] The target display module 430 is configured to, in response to the target detection result representing that there is a target object in the complete target package, perform display processing on the target object.

[0086] ​In the exemplary imaging display device based on the security inspection machine, the application buffers the image rendered by the scanned ray data when the security inspection machine scans the package, performs complete package identification processing on the obtained current buffer image to obtain a complete package identification result, and determines whether complete package information exists in the current buffer image. If the complete package identification result indicates that complete inspection package exists in the current buffer image, target detection processing is performed on the package image region where the complete inspection package is located in the current buffer image to obtain a target detection result. It is determined whether the complete inspection package has a target object that needs to be focused on. In response to the target detection result indicating that the target object exists in the complete inspection package, the target object is displayed. Compared with the prior art, the current scanned ray data is rendered and displayed in real time when the security inspection machine scans the package, and the real-time image pushing and fast moving display will cause eye fatigue. The method of the application can adjust the data pushing mode from one or more groups of ray data to complete package data, omit the real-time pushing process, and reduce eye fatigue. The method of the application further displays the package with the target object, reduces the number of images that need to be judged by the security inspector, and reduces the security inspection pressure.

[0087] It should be noted that the apparatus provided in the above embodiments and the method provided in the above embodiments belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiments, which will not be repeated here. The apparatus provided in the above embodiments can be divided into different functional modules according to the needs, i.e., the internal structure of the apparatus is divided into different functional modules to complete all or part of the functions described above, which is not limited here.

[0088] The functions of each module can be referred to the embodiment of the imaging display method based on the security inspection machine, which will not be repeated here.

[0089] Please refer to Figure 5 , Figure 5 is a structural schematic diagram of an embodiment of the electronic device. The electronic device 100 includes a memory 101 and a processor 102, and the processor 102 is configured to execute program instructions stored in the memory 101 to implement the steps in any of the above imaging display methods based on the security inspection machine. In one specific implementation scenario, the electronic device 100 can include but is not limited to a microcomputer, a server, and in addition, the electronic device 100 can also include a notebook computer, a tablet computer and other mobile devices, which are not limited here.

[0090] Specifically, the processor 102 is configured to control itself and the memory 101 to implement the steps in any of the above imaging display method embodiments based on the security inspection machine. The processor 102 can also be referred to as a CPU (Central Processing Unit). The processor 102 can be an integrated circuit chip with processing capability. The processor 102 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. In addition, the processor 102 can be implemented by an integrated circuit chip together.

[0091] In the example electronic device, the present application buffers the image rendered by the scanned ray data when the security inspection machine scans the package, performs complete package identification processing on the obtained current buffer image to obtain a complete package identification result, and determines whether complete package information exists in the current buffer image. If the complete package identification result indicates that complete package information exists in the current buffer image, target detection processing is performed on the package image region where the complete package exists in the current buffer image to obtain a target detection result. It is determined whether the complete package has a target object that needs to be focused on. In response to the target detection result indicating that the complete package has a target object, display processing is performed on the target object. Compared with the prior art, the current scanned ray data is rendered and displayed in real time when the security inspection machine scans the package, and real-time image pushing and fast moving display can cause eye fatigue. The method of the present application can adjust the data pushing mode from one or more groups of ray data to complete package data, omit the real-time pushing process, and reduce eye fatigue. The method of the present application further displays the package with the target object, reduces the number of images that need to be judged by the security inspector, and reduces the security inspection pressure.

[0092] Please refer to Figure 6 , Figure 6 is a structural schematic diagram of an embodiment of the computer readable storage medium of the present application. The computer readable storage medium 110 stores program instructions 111 capable of being executed by the processor, and the program instructions 111 are used to implement the steps in any of the above imaging display method embodiments based on the security inspection machine.

[0093] In the exemplary storage medium, by running the program instructions in the storage medium, the image rendered by the scanned ray data when the security inspection machine scans the package is buffered first, the complete package identification result is obtained by performing complete package identification processing on the obtained current buffer image, and whether complete package information exists in the current buffer image is judged. If the complete package identification result represents that complete inspection package exists in the current buffer image, target detection processing is performed on the package image area where the complete inspection package is located in the current buffer image, and a target detection result is obtained. Whether the complete inspection package has a target object that needs to be focused on is judged. In response to the target detection result representing that the target object exists in the complete inspection package, the target object is displayed. Compared with the prior art, when the security inspection machine scans the package, the current scanned ray data is rendered and displayed in real time, and the real-time image pushing and fast moving display will cause eye fatigue. The method of the application can adjust the data pushing mode from one or more groups of ray data pushing to complete package data pushing, omit the real-time pushing process, and reduce eye fatigue. The method of the application further displays the package with the target object, reduces the number of images that need to be judged by the security inspection personnel, and reduces the security inspection pressure.

[0094] In some embodiments, the apparatus provided by the embodiments of the present disclosure has functions or includes modules that can be used to perform the methods described in the above method embodiment descriptions, and the specific implementation can refer to the descriptions of the above method embodiments. For brevity, they will not be repeated here.

[0095] The above description of various embodiments tends to emphasize the differences between various embodiments, and the same or similar parts can be mutually referred to. For brevity, they will not be repeated here.

[0096] In several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the above-described device implementation is only schematic, for example, the division of modules or units is only a logical function division, and actual implementation can have another division manner, for example, a unit or component can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed mutual ones can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0097] In addition, each of the function units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit. When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application, essentially or in the form of a contribution to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the methods in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

Claims

1. An imaging display method based on a security inspection machine, characterized in that, The method includes: The acquired current ray data is rendered and imaged to obtain the current rendered image; the current rendered image is stored in a preset buffer to obtain the current buffered image. Perform complete package recognition processing on the current buffer image to obtain the complete package recognition result; In response to the complete package recognition result indicating the presence of a complete package to be inspected in the current buffer image, target detection processing is performed on the package image region where the complete package to be inspected is located to obtain the target detection result; and the current buffer image in the preset buffer is cleared. In response to the target detection result indicating the presence of a target object in the complete package to be inspected, the target object is displayed. If the target detection result indicates that the target object does not exist in the complete package to be inspected, then the complete package to be inspected will not be displayed. In response to the complete package recognition result indicating the presence of an incomplete package to be inspected in the current buffer image, it is determined whether the number of current ray data is greater than a ray number threshold or whether the number of current rendered images is greater than a frame number threshold; if so, image features of the incomplete package to be inspected are extracted to obtain local features; target prediction processing is performed based on the local features to obtain at least one predicted object and the prediction probability corresponding to each predicted object; the predicted object is displayed based on the prediction probability. The frequency of occurrence of the target object within the preset period is determined based on the target detection results within the preset period; the speed of the conveyor belt of the security inspection machine is adjusted according to the frequency of occurrence, wherein the frequency of occurrence is negatively correlated with the speed of the conveyor belt.

2. The method according to claim 1, characterized in that, After performing complete package recognition processing on the acquired current buffer image to obtain the complete package recognition result, the method further includes: In response to the complete package recognition result indicating that there is an incomplete package to be inspected in the current buffer image, the continuously acquired current ray data is subjected to rendering imaging processing to obtain the current rendered image; The current buffer image is updated based on the current rendered image until the complete package recognition result of the updated current buffer image indicates that the complete package to be inspected exists in the updated current buffer image; The target detection process is performed on the package image region where the complete package to be inspected is located to obtain the target detection result.

3. The method according to claim 2, characterized in that, The current buffered image is cached in a buffer, and the buffer update process for the current buffered image based on the current rendered image includes: Buffer the currently rendered image into the buffer; The current buffer image is updated based on the current rendered image in the buffer to obtain the updated current buffer image.

4. The method according to claim 1, characterized in that, The step of performing target detection processing on the package image region where the complete package to be inspected is located, to obtain the target detection result, includes: Obtain the preset image features of each target object; Image feature extraction processing is performed on the complete package to be inspected to obtain the features to be inspected; Based on the feature similarity between the feature to be detected and each preset feature, the target feature with a feature similarity greater than a similarity threshold is determined from each preset feature; In response to the fact that the number of the target features is not zero, a target detection result is obtained that indicates the presence of the target object in the complete package to be inspected; When the number of the target features is zero, a target detection result is obtained that indicates that the target object does not exist in the complete package to be inspected.

5. The method according to claim 1, characterized in that, The display processing of the target object includes: Based on the position information of the target object in the current buffer image, the current buffer image is processed by image matting to obtain the image of the target object; The target object image is subjected to image optimization processing to obtain an optimized image; The optimized image is then displayed.

6. The method according to claim 5, characterized in that, The step of performing image optimization processing on the target object image to obtain an optimized image includes: Obtain the material type of the target object; In response to the fact that the material type of the target object is inorganic, the organic matter in the target object image is determined based on the ray data corresponding to the target object image; Organic matter is removed from the target object image to obtain the optimized image.

7. An electronic device, characterized in that, It includes a memory and a processor, the processor being configured to execute program instructions stored in the memory to implement the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.

Images